Blood biomarkers of long COVID: A cross-sectional evaluation of key cytokines and the coordination of inflammation
5 affiliations
- 1School of Health, Sport and Rehabilitation, College of Health and Humanities, University of Derby, Derby, United Kingdom.
- 2Research Centre for Healthcare and Communities Transformation, Coventry University, Coventry, United Kingdom.
- 3Department of Physiological Sciences, Faculty of Science, Stellenbosch University, Stellenbosch, South Africa.
- 4Department of Economics, Faculty of Economic and Management Sciences, Stellenbosch University, Stellenbosch, South Africa.
- 5School of Sport, Exercise and Health Sciences, Loughborough University, Loughborough, United Kingdom.
Abstract
Introduction: Long COVID is a complex pathophysiological condition linked to a heterogenous symptom profile. Dysregulation of 1) immunological and inflammatory responses, 2) endothelial/vascular function, and 3) metabolic and clotting processes have been identified; however, there remains a lack of clinically validated diagnostic biomarkers. The study aim was to investigate dysregulation of cytokines, chemokines and growth factors in confirmed Long COVID cases grouped by disease severity. Methods: One-hundred-and-twenty-seven participants provided venous blood samples for cytokine profiling utilising Luminex assays. Participants were grouped into those with no knowledge of SARS-CoV-2 infection (n = 21), acute infection within 3-months (n = 22), spontaneous resolvers (>3-months, n = 21) and those with confirmed Long COVID (n = 63). Long COVID severity was grouped according to the post-COVID-19 functional status (PCFS) and defined as those with negligible (4A; n = 9), mild (4B; n = 20), moderate (4C; n = 24), and severe limitations (4D; n = 10). Cytokines were assessed for normality using Shapiro-Wilk tests and differences measured between groups using Kruskal-Wallis tests. Pairwise correlation and principal component analysis were performed. Results: Interleukin-12 (p40/p70) was 22% greater in the Long COVID group (Median [IQR]: 58 [49] pg/mL, p = 0.046) and 30% greater in the acutely infected group (64 [36] pg/mL; p = 0.039) against non-infected controls (45 [25] pg/mL). Following PCFS classification, fibroblast growth factor-basic was 25% lower in Long COVID 4B versus spontaneous resolvers (24 [7] vs. 32 [24] pg/mL; p = 0.025). Pairwise correlation differences demonstrated that three Long COVID subgroups had strong, positive associations between cytokines when compared to non-infected controls; however, PCA was unable to determine any coordinated cytokine variation that was group specific. Discussion and conclusions: Elevated interleukin-12 (p40/p70) and reduced basic fibroblast growth factor suggest persistent immune activation and possible subgroup-specific Long COVID; however, these findings were marginally significant and exploratory. A coordinated but partial immunological/inflammatory profile may exist in some of those with Long COVID, suggesting a rationale for subgrouping upon a biological, rather than self-report basis but larger longitudinal studies are required to confirm this hypothesis.
Related
- Modulating Th17-driven inflammation in ulcerative colitis with Chinese medicine: pathways, cytokines and beyond.
- Development of immunological reagents for the identification of biomarkers of inflammation in the common marmoset
- Selected interleukins, soluble TNF-α receptors, c-reactive protein and cortisol as biomarkers of surgical trauma-induced inflammation in children.
- Sequencing based dissection of inflammation associated natural killer cell remodeling in lung adenocarcinoma identifies state biomarkers and therapeutic vulnerabilities
- Fecal biomarkers to characterize intestinal inflammation among children with medically attended diarrhea across six low-resource settings: Findings from the Enterics for Global Health (EFGH) Shigella surveillance study
BibTeX
@article{thomas2026blood,
title = {{Blood biomarkers of long COVID: A cross-sectional evaluation of key cytokines and the coordination of inflammation}},
author = {Callum Thomas and Ruth E Ashton and Etheresia Pretorius and Rebecca Owen and Massimo Nunes and Jan H Pretorius and Isaac T Shawa and Corinna Chidley and Mark A Faghy},
journal = {PLOS One},
year = {2026},
month = oct,
doi = {10.1371/journal.pone.0360525},
url = {https://doi.org/10.1371/journal.pone.0360525},
note = {CC BY},
}
RIS
TY - JOUR
TI - Blood biomarkers of long COVID: A cross-sectional evaluation of key cytokines and the coordination of inflammation
AU - Callum Thomas
AU - Ruth E Ashton
AU - Etheresia Pretorius
AU - Rebecca Owen
AU - Massimo Nunes
AU - Jan H Pretorius
AU - Isaac T Shawa
AU - Corinna Chidley
AU - Mark A Faghy
PY - 2026
DA - 2026/10/21
JO - PLOS One
DO - 10.1371/journal.pone.0360525
UR - https://doi.org/10.1371/journal.pone.0360525
AB - Introduction: Long COVID is a complex pathophysiological condition linked to a heterogenous symptom profile. Dysregulation of 1) immunological and inflammatory responses, 2) endothelial/vascular function, and 3) metabolic and clotting processes have been identified; however, there remains a lack of clinically validated diagnostic biomarkers. The study aim was to investigate dysregulation of cytokines, chemokines and growth factors in confirmed Long COVID cases grouped by disease severity. Methods: One-hundred-and-twenty-seven participants provided venous blood samples for cytokine profiling utilising Luminex assays. Participants were grouped into those with no knowledge of SARS-CoV-2 infection (n = 21), acute infection within 3-months (n = 22), spontaneous resolvers (>3-months, n = 21) and those with confirmed Long COVID (n = 63). Long COVID severity was grouped according to the post-COVID-19 functional status (PCFS) and defined as those with negligible (4A; n = 9), mild (4B; n = 20), moderate (4C; n = 24), and severe limitations (4D; n = 10). Cytokines were assessed for normality using Shapiro-Wilk tests and differences measured between groups using Kruskal-Wallis tests. Pairwise correlation and principal component analysis were performed. Results: Interleukin-12 (p40/p70) was 22% greater in the Long COVID group (Median [IQR]: 58 [49] pg/mL, p = 0.046) and 30% greater in the acutely infected group (64 [36] pg/mL; p = 0.039) against non-infected controls (45 [25] pg/mL). Following PCFS classification, fibroblast growth factor-basic was 25% lower in Long COVID 4B versus spontaneous resolvers (24 [7] vs. 32 [24] pg/mL; p = 0.025). Pairwise correlation differences demonstrated that three Long COVID subgroups had strong, positive associations between cytokines when compared to non-infected controls; however, PCA was unable to determine any coordinated cytokine variation that was group specific. Discussion and conclusions: Elevated interleukin-12 (p40/p70) and reduced basic fibroblast growth factor suggest persistent immune activation and possible subgroup-specific Long COVID; however, these findings were marginally significant and exploratory. A coordinated but partial immunological/inflammatory profile may exist in some of those with Long COVID, suggesting a rationale for subgrouping upon a biological, rather than self-report basis but larger longitudinal studies are required to confirm this hypothesis.
SN - 1932-6203
ER -